1. Field of the Invention
The present invention relates to an image generation system for generating a three-dimensional mixed spatial image by fusing a real-space image and a virtual-space image, and the like.
2. Description of the Related Art
Recently, in the field of various types of image processing, a technique of synthesizing a virtual-space image, such as CG (computer graphics) or the like, with a real-space image (photographed image) by fusing a real space and a virtual space is being used.
When intending to fuse a real space and a virtual space, image synthesis of an object in the real space and an object in the virtual space causes problems. As shown in FIG. 9, when intending to simply paste an object in a virtual space 901 making a real space 902 a background, the object in the virtual space 901 is always displayed in front of objects in the real space 902 (see a fused space 903).
However, an object in a virtual space is sometimes positioned behind an object in a real space. In such a case, it is necessary to display the object in the virtual space behind the object in the real space.
A system disclosed in Japanese Patent Application Laid-Open (Kokai) No. 11-331874 (1999), serving as a conventional technique for knowing the position of a moving object in a real space in real time, obtaining a fore and aft relationship with an object in a virtual space, and displaying an image which is consistent in the fore and aft relationship, includes a position sensor for time serially inputting standing-point-position information, a stereoscopic camera where a plurality of images are consecutively input time serially, and an image processing apparatus.
The image processing apparatus detects depth images from stereoscopic images, which are input in consecutive time series, assumes the observer""s standing point in a future time at which the observer is provided with a three-dimensional image, based on a change in the standing point in the past which has been input from the position sensor, and consecutively changes consecutively obtained depth images into images from the assumed standing point in the future.
However, in the system disclosed in the above-described patent application, since the standing point is assumed only based on information relating to a change in the standing point in the past which has been time serially input from the position sensor, accuracy in assumption is insufficient from the viewpoint of providing a more precise composite real space.
The present invention has been made in consideration of the above-described problems.
It is an object of the present invention to provide a composite-reality providing system in which, when fusing a real space and a virtual space, by exactly knowing the positional relationship between an object in the real space and an object in the virtual space, a very precise three-dimensionally consistent mixed real space can be provided.
According to one aspect, the present invention which achieves the above-described object relates to an image-display control apparatus including detection means for detecting a predetermined object from a photographed image, position calculation means for obtaining a positional relationship between the predetermined object in the photographed image detected by the detection means and an object in a virtual image, image removing means for removing a predetermined partial image in the object in the virtual image by referring to a result of calculation of the position calculation means, synthesis means for synthesizing the virtual image and the photographed image so that the predetermined object in the photographed image is present at a position of the partial image in the virtual image removed by the image removing means, and display control means for displaying an image obtained as a result of synthesis of the synthesis means on display means.
According to another aspect, the present invention which achieves the above-described object relates to an image-display control apparatus including detection means for detecting a predetermined object from each of first and second photographed images obtained by photographing a real space, position calculation means for obtaining a positional relationship between the predetermined objects in the first and second photographed images detected by the detection means and an object in a virtual image, image removing means for removing predetermined partial images in the object in the virtual image by referring to a result of calculation of the position calculation means, synthesis means for forming a first synthesized image by synthesizing the virtual image and the first photographed image and forming a second synthesized image by synthesizing the virtual image and the second photographed image so that the predetermined objects in the first and second photographed images are present at respective positions of the partial images in the virtual image removed by the image removing means, and display control means for displaying the first synthesized image obtained by the synthesis means on display means for a left eye, and displaying the second synthesized image on display means for a right eye.
According to still another aspect, the present invention which achieves the above-described object relates to an image processing system including a camera for photographing a real space, a position/direction sensor for detecting a position and a direction of an object in the real space, a computer for synthesizing the photographed image photographed by the camera and a virtually formed virtual image, and display means for displaying an image obtained as a result of the synthesis by the computer. The computer includes image input means for inputting the photographed image photographed by the camera, detection means for detecting a predetermined object from the photographed image, position calculation means for obtaining a positional relationship between the predetermined object in the photographed image detected by the detection means and an object in a virtual image, image removing means for removing a predetermined partial image in the object in the virtual image by referring to a result of the calculation of the position calculation means, synthesis means for synthesizing the virtual image and the photographed image so that the predetermined object in the photographed image is present at a position of the partial image in the virtual image removed by the image removing means, and display control means for displaying an image obtained as a result of the synthesis of the synthesis means on the display means.
According to yet another aspect, the present invention which achieves the above-described object relates to an image generation system including first and second cameras for photographing the same real space, a position/direction sensor for detecting a position and a direction of an object in the real space, a computer for outputting first and second synthesized images by synthesizing first and second photographed images photographed by the first and second cameras, respectively, and a virtually formed virtual image, and display means for a right eye and display means for a left eye for displaying the first and second synthesized images, respectively. The computer includes image input means for inputting the first and second photographed images, detection means for detecting respective predetermined objects from the first and second photographed images, position calculation means for obtaining a positional relationship between the predetermined objects in the first and second photographed images detected by the detection means and an object in a virtual image, image removing means for removing predetermined partial images in the object in the virtual image by referring to a result of calculation of the position calculation means, synthesis means for forming a first synthesized image by synthesizing the virtual image and the first photographed image and forming a second synthesized image by synthesizing the virtual image and the second photographed image so that the predetermined objects in the first and second photographed images are present at respective positions of the partial images of the virtual image removed by the image removing means, and display control means for displaying the first synthesized image obtained by the synthesis means on display means for a left eye, and displaying the second synthesized image on display means for a right eye.
According to still another aspect, the present invention which achieves the above-described object relates to an image-display control method including a detection step of detecting a predetermined object from a photographed image, a position calculation step of obtaining a positional relationship between the predetermined object in the photographed image detected in the detection step and an object in a virtual image, an image removing step of removing a predetermined partial image in the object in the virtual image by referring to a result of the calculation in the position calculation step, a synthesis step of synthesizing the virtual image and the photographed image so that the predetermined object in the photographed image is present at a position of the partial image in the virtual image removed in the image removing step, and a display control step of displaying an image obtained as a result of synthesis in the synthesis step on display means.
According to still another aspect, the present invention which achieves the above-described object relates to an image-display control method including a detection step of detecting a predetermined object from each of first and second photographed images obtained by photographing the same real space, a position calculation step of obtaining a positional relationship between the predetermined objects in the first and second photographed images detected in the detection step and an object in a virtual image, an image removing step of removing predetermined partial images in the object in the virtual image by referring to a result of calculation in the position calculation step, a synthesis step of forming a first synthesized image by synthesizing the virtual image and the first photographed image, and forming a second synthesized image by synthesizing the virtual image and the second photographed image so that the predetermined objects in the first and second photographed images are present at respective positions of the partial images of the virtual image removed in the image removing step, and a display control step of displaying the first synthesized image obtained in the synthesis step on display means for a left eye, and displaying the second synthesized image on display means for a right eye.
The foregoing and other objects, advantages and features of the present invention will become more apparent from the following detailed description of the preferred embodiments taken in conjunction with the accompanying drawings.